THERMOS: Thermally-Aware Multi-Objective Scheduling of AI Workloads on Heterogeneous Multi-Chiplet PIM Architectures

TL;DR

THERMOS is a reinforcement learning-based scheduling framework that optimizes AI workload performance and energy efficiency on heterogeneous chiplet PIM architectures while considering thermal constraints.

Contribution

It introduces a thermally-aware, multi-objective reinforcement learning approach for scheduling AI workloads on heterogeneous multi-chiplet PIM systems, enabling Pareto-optimal trade-offs.

Findings

Achieves up to 89% faster execution time.

Reduces energy consumption by 57%.

Maintains minimal runtime and energy overhead.

Abstract

Chiplet-based integration enables large-scale systems that combine diverse technologies, enabling higher yield, lower costs, and scalability, making them well-suited to AI workloads. Processing-in-Memory (PIM) has emerged as a promising solution for AI inference, leveraging technologies such as ReRAM, SRAM, and FeFET, each offering unique advantages and trade-offs. A heterogeneous chiplet-based PIM architecture can harness the complementary strengths of these technologies to enable higher performance and energy efficiency. However, scheduling AI workloads across such a heterogeneous system is challenging due to competing performance objectives, dynamic workload characteristics, and power and thermal constraints. To address this need, we propose THERMOS, a thermally-aware, multi-objective scheduling framework for AI workloads on heterogeneous multi-chiplet PIM architectures. THERMOS trains a single multi-objective reinforcement learning (MORL) policy that is capable of achieving Pareto-optimal execution time, energy, or a balanced objective at runtime, depending on the target preferences. Comprehensive evaluations show that THERMOS achieves up to 89% faster average execution time and 57% lower average energy consumption than baseline AI workload scheduling algorithms with only 0.14% runtime and 0.022% energy overhead.